Exercise 1 - Using the select function

Using the select function pulls columns that are selected from a data frame. Using storms as the source, the numerical values can be grabbed the.

or the character values can be used.

The columns can even be selected based on looking for certain conditions, specifically here columns with diameter in their name

Using the %like% function from the data.table library, partial string matches within the columns can be selected. Here, tropical is used to select both tropical depression and tropical storm

Exercise 2 - mutate() across() everything(). “Everything, sir?”, “EVERYTHING!!!”

Continuing with the storms data, the mutate() function combined with across() and everything() change the values of the data frame to make everything the same type. In the example below, the storms data is mutated using the mutate() function. Where the changes are set to be done across the whole data frame using across(), and using every data type is selected by everything(). The data type the selected to mutate into is character or chr.

The where() function can be used to convert data of a specific type, where you might want to keep boolean variables or coordinates, ord, or convert integers to floats. Here the numerals are being converted to characters.

Exercise 3 - arrange() and count()

The count() function is used on storms to count the number of elements within year and weighted by name. This returns a table ordered by year. The arrange() function, with desc(), can be used to change the order of the storms to give the longest active storm by how frequently a storm name occurs in the the list by the count function.

To counter this, the unique() function will also be used to remove duplicate counts where name and year are both the same. (Though there may still be an error where a storm continues over the new year and it may be counted twice.). Using unique() removes the requirement to weight the storm by name in count().

Exercise 4 plotting schemes results

The select() and filter() functions will be used to select data to plot with ggplot() from the storms dataset. For this the Tomas dataset will be selected from table (because it is a most excellent name for a storm!). The values for name, wind speed, pressure, and status are selected. Category was requested which gives a numeral value of the of the status of the storm but using status is a bit more informative for the layman.

Plotting more than one storm and using colour to define the different storms.

Bonus Round

Plotting the path of Tomas and Clara using GPS coordinates. Tomas is in Green and Clara in Red. (I haven’t yet worked out the addLegend function yet…)